“Biochip” is the generic term for devices in which biological substances that chemically react with to-be-detected biological substances in a specific manner are fixed at predetermined positions on a chip surface.
A DNA chip that is a typical example of the biochip is used to detect the types and amounts of target DNA included in blood or cell extract.
The DNA chip has, for example a structure in which thousands to tens of thousands of types of probe DNA, each being single-chain DNA having a known sequence, are arranged in an array on a substrate such as a glass slide.
When a to-be-examined liquid containing fluorescence-marked target DNA is supplied to the DNA chip, only the target DNA which has sequences complementary to the sequences of the probe DNA is fixed by the complementary sequences of target DNA and probe DNA hydrogen-bonding to each other and forming a double chain. As a result, the parts to which the target DNA is fixed is fluorescent-colored. By measuring the position and coloring intensity of the fluorescent-colored parts on the chip, the types and amounts of the target DNA can be detected.
In order to make a DNA chip used in this manner, it is necessary to fix a plurality types of probe DNA each having a predetermined sequence at predetermined positions on a substrate surface.
There are two main methods of fixing probe DNA. A first method is called “micro array method”. In this method, probe DNA that has been chemically synthesized or extracted from an intended living organism in advance is fixed in an array on a substrate by dropping or printing.
In a second method, using four bases thymine (T), adenine (A), cytosine (C) and guanine (G), a plurality of types of probe DNA, each being single-chain DNA having a predetermined sequence according to design, are chemically synthesized directly on a substrate.
In the second fixing method, it is required that on a surface of a substrate used for the chemical synthesis, a reaction region where a reaction to synthesize intended probe DNA proceeds and a non-reaction region which is not involved in the synthesis reaction should be formed in a definitely separated manner.
As a DNA chip made by this second fixing method, for example “Gene Chip” (name of an article produced by Affymetrix, Inc.) is known.
In the case of this chip, quartz is used as a material for a substrate, and photolithography is applied to form a reaction region and a non-reaction reaction in a separated manner. Further, when probe DNA is chemically synthesized, single-chain DNA is formed unit by unit by applying ultraviolet rays for activation of single-chain DNA in the reaction region.
In this chip, a reaction region consisting of about 200,000 spots of about 20 μm square each is formed on a single substrate, and about 2,000,000 strands of probe DNA of the same type are fixed in one spot.
In this chip, the spots are formed on the chip at high density. Hence, a very large number of types of target DNA can be detected in one test.
FIG. 2A of PCT Application Published Japanese Translation No. Hei 9-500568 shows an array plate described below.
The array plate is made as follows: First, a surface of an Si substrate is made to react with fluoroalkylsilane to once form a thin film of hydrophobic fluoroalkylsiloxane on it. Then, this thin film is removed in a predetermined two-dimensional pattern so that the surface of the Si substrate will be exposed in the spots having the thin film removed. Last, the exposed surface is made to react with hydroxysilane or alkylsilane so that the exposed surface will have OH group.
Thus, in this array plate, the surface of the Si substrate has sites of a hydrophobic thin film having a large surface tension and sites having hydrophilic OH group. For biological substances, the former function as a non-reaction region, while the latter function as a reaction region.
When this array plate is used, synthesis reaction is made to proceed in the hydrophilic sites, and then a to-be-examined liquid is supplied to these sites. Due to the large surface tension of the hydrophobic thin film around the hydrophilic sites, the to-be-examined liquid is held in the hydrophilic sites.
In this array plate, however, the reaction region and the non-reaction region are formed on the surface of the Si substrate to be virtually flush with each other. Hence it does not have sufficient stability in holding a supplied to-be-examined liquid and hence is not easy to use.
Further, the formation of the reaction region and the formation of the non-reaction region both depend on chemical reaction between the Si surface and other chemicals. The reaction does not always proceed at a yield of 100%. Thus, it is not improbable that the boundary between the reaction region and the non-reaction region is indefinite. Another problem is that the hydrophobic thin film and the hydrophilic sites are easily damaged from the outside.
From the aspect of holding an to-be-examined liquid, a substrate in which bottomed wells having a structure capable of holding a to-be-examined liquid are distributed in a substrate surface is preferable to the array plate having the above structure.
As a substrate of this type, PCT Application Published Japanese Translation No. 2002-537869 discloses a substrate used for direct synthesis of probe DNA and a method of chemical synthesis of probe DNA using it.
A sketch of this substrate is given in FIG. 1.
The substrate 1 is made from an Si wafer. In a surface of the substrate 1, a plurality of micro cuvettes 2 are formed in a predetermined array. The micro cuvettes 2 are sunken holes (bottomed wells) of about 1 to 1000 μm in diameter and about 1 to 500 μm in depth, and function as a reaction region for chemical synthesis of probe DNA. The surface part except for these wells is a non-reaction region.
This substrate A is made as follows:
Step a1: By applying photolithography and etching to a surface 1a of an Si wafer 1, an intermediate A1 as shown in FIG. 2 is formed, in which sunken holes 2A of almost the same shape as that of to-be-formed bottomed wells are formed at positions at which the bottomed wells are to be formed.
Step a2: By performing, for example thermal oxidation treatment onto the surface 1a of the Si wafer and the surfaces of the sunken holes 2A (their bottoms and side surfaces) in the intermediate A1, an intermediate A2 as shown in FIG. 3 is formed, in which only the superficial part of these surfaces has been turned into an Si oxide layer 3 of about 0.5 μm in thickness.
Step a3: Silanization treatment is performed onto the surface of the Si oxide layer 3 of the intermediate A2. Specifically, the surface of the Si oxide layer 3 is treated with alkali by applying Brown Process using NaOH, and then treated with, for example an activated silanizing agent having an epoxysilane type group. Then, linkage of the silanizing agent and hydrolysis of epoxy resin is made to proceed successively. As a result, an intermediate A3 shown in FIG. 4 is obtained, in which a silane layer 4 is formed on the surface of the Si oxide layer 3.
In the intermediate A3, since OH groups of silane are present in its entire surface, the entire surface can react with DNA phosphoramidite.
Step a4: By applying the phosphoramidite method to the surface of the intermediate A3, strands of single-chain DNA of a length corresponding to 5 bases or so are synthesized using DNA phosphoramidite T (thymine). Thus, an intermediate A4 having bottomed wells as shown in FIG. 5 is formed, in which an oligonucleotide (5T) spacer layer 5 is formed on the silane layer 4.
It is to be noted that the terminal of the synthesized oligonucleotide (5T) is blocked with dimethoxytrityl (DMT).
The entire surface of the intermediate A4 is covered with the spacer layer 5 of oligonucleotides (5T) blocked with DMT. Hence, when the terminals of oligonucleotides are activated by eliminating the DMT blocking the terminals (detrirylation), probe DNA can be synthesized there.
Thus, in the case of the intermediate A4, when a DNA chip is to be made, the entire surface formed of the oligonucleotide (5T) spacer layer 5 functions as a reaction region.
However, this does not meet the condition for the substrate for synthesizing probe DNA, namely the condition that the bottomed wells and the other part should be definitely separated as a reaction region and a non-reaction region, respectively.
Hence, treatment needs to be performed on the layer 5 of the intermediate A4 to leave the bottomed wells as they are a reaction region and turn the other part into a non-reaction region. This treatment is capping performed in the next step.
Step a5: As shown in FIG. 6, in the intermediate A4, only the bottomed wells are filled with resin droplets 6. In this state, capping is performed on the layer 5.
Specifically, by performing detritylation on the oligonucleotide (5T) spacer layer 5, the terminals of the oligonucleotides (5T) are activated. Then, using trichloroacetic acid, acetic anhydride, dymethylaminopyridine or the like, the activated terminals of the oligonucleotides (5T) are blocked and inactivated.
Then, using an organic solvent such as tetrahydrofuran, the resin droplets 6 filling the bottomed wells are dissolved and removed so that the layer 5 will be exposed in the bottomed wells.
During this capping, since the bottomed wells are filled with resin, the oligonucleotide (5T) spacer layer 5 in the bottomed wells does not undergo the capping and maintains the state capable of reaction. Meanwhile, the oligonucleotides (5T) in the part except for the bottomed wells undergo the capping and are brought into a state incapable of synthesis reaction.
In this way, a substrate A for synthesis of probe DNA having a cross-sectional structure shown in FIG. 7 is made.
In this substrate A, the bottomed wells 2, which are sunken holes, are formed in the surface of the Si wafer 1 in a predetermined pattern. The Si oxide layer 3 is formed on the Si wafer 1 to cover the entire surface thereof, and the silane layer 4 is formed on the Si oxide layer 3 to cover the entire surface thereof.
On the bottom 2a and the side surface 2b of each of the bottomed wells 2, the spacer layer 5 of oligonucleotides (5T) with their terminals blocked with DMT is exposed. These spots form a reaction region for synthesizing probe DNA. Meanwhile, the part 5a except for these spots of the layer 5 has undergone capping and forms a non-reaction region.
When a DNA chip is made using this substrate A by the phosphoramidite method, chemical synthesis of probe DNA proceeds in the bottomed wells 2.
However, this substrate A has problems mentioned below.
A first problem is that in the substrate A, the boundary between the reaction region and the non-reaction region is determined by how the bottomed wells are filled with resin droplets in step a5.
Generally, filling of the bottomed wells with resin droplets is performed using a piezoinjector. The amount of resin droplets supplied to fill one bottomed well is very minute, specifically in the order of p1 to μ1. Hence, in step a5, sometimes the amount of resin droplets supplied to fill a bottomed well is too much and the resin runs over the bottomed well, and sometimes the amount of resin droplets supplied is too less to completely fill a bottomed well.
In the former case, also a surface of a part surrounding the bottomed well is covered with the resin running over the bottomed well. As a result, in the capping performed next, the surface of this surrounding part does not undergo capping and remains capable of synthesis reaction.
Hence, when a DNA chip is made, probe DNA is chemically synthesized also on the surface of the part surrounding the bottomed well. As a result, when target DNA is examined using the DNA chip made, fluorescent coloring may occur not only in the bottomed well but also in the part surrounding the bottomed well over which the resin ran. This hinders accurate reading of fluorescent marks.
Further, in the case in which the amount of resin droplets supplied is too less to fill a bottomed well, the thickness of the resin covering the inside surface of the bottomed well is thin. Hence, the resin is easily corroded by an acid solution used in capping. As a result, a part of the oligonucleotide (5T) spacer layer located in the bottomed well may undergo capping.
Thus, when a DNA chip is made, probe DNA may not be satisfactorily chemically synthesized in this bottomed well. As a result, when target DNA is examined, fluorescent coloring may not occur with a sufficient coloring intensity in this bottomed well.
Further, if the capping in step a5 is insufficient, it also may cause the problem that when a DNA chip is made and used, fluorescent coloring occurs not only in bottomed wells but also in parts where capping was insufficient. In other words, background noise easily occurs.
Further, when the sunken holes 2A, which will form the bottomed wells, are formed in step a1, the depth of the sunken holes 2A is determined by the length of etching time. Hence, if time management for etching is not performed accurately, the bottomed wells may not be formed to have an accurate depth according to design criteria.